388

Waterbird-mediated Productivity of Two Soda Pans in the Carpathian Basin in Central Europe

Emil Boros1,*, CsaBa PigniCzki2, Tamás sáPi2, kaTalin V.-Balogh1, lajos Vörös1 and Boglárka somogyi1

1MTA Centre for Ecological Research, Balaton Limnological Institute, P.O. Box 35, H-8237 Tihany, Hungary

2Kiskunság National Park Directorate, P.O. Box 186, H-6000 Kecskemét, Hungary

*Corresponding author; E-mail: boros.emil@okologia.mta.hu

Abstract.—The effect of aquatic birds on nutrient cycling and energy flow was investigated in two soda pans, one turbid and the other colored, with different physical and chemical characteristics. Primary plankton produc-tion and respiration were measured together with an estimation of waterbird carbon, nitrogen and phosphorus loading during 2014. Both pans were hypertrophic and showed net heterotrophy. The nutrient loading of the birds in the turbid pan was approximately five times higher (C: 758 kg/ha/year, N: 122 kg/ha/year, P: 20 kg/ha/year) than in the colored pan, with significant guanotrophication. Despite the high chlorophyll a concentrations (turbid: 752 µg/l and colored: 369 µg/l, on average), the annual surface-related planktonic production was relatively low (turbid: 64 mg C/m2/year and colored: 23 mg C/m2/year), by contrast, respiration was similar in the two pans (turbid: 75 C/m2/year and colored: 78 mg C/m2/year). Nutrient loading showed a significant positive correlation with total and soluble reactive phosphorus, chlorophyll a and gross planktonic production, supporting the conclu-sion that the waterbirds significantly affected primary production. By contrast, there was no significant correlation between the nutrient loading and planktonic respiration. The low production and respiration ratio (Pro/Res) in the colored pan was presumably caused by a high dissolved organic carbon concentration (polyhumic). A possible explanation for the difference of Pro/Res between the turbid and colored pans is variation in the decomposition of the bird excrement and surrounding macrophytes. Received 20 January 2016, accepted 22 August 2016.

Key words.—colored pan, guanotrophication, heterotrophy, Ramsar sites, respiration, turbid pan.Waterbirds 39(4): 388-401, 2016

The role of waterbirds in biochemical cycles and productivity in aquatic ecosys-tems has been studied in a number of cases. Guanotrophication by waterbirds can have a positive impact on productivity and energy flow. However, waterbirds can cause local habitat degradation. Several supporting and regulatory ecosystem services are provided by waterbirds (Green and Elmberg 2013). The influence of birds in aquatic ecosystems is often indirect, and their importance re-mains largely unknown because of the lack of essential information about the interac-tions between waterbirds and other ecosys-tem components. Thus, effective manage-ment of aquatic ecosystems requires a better understanding of how waterbirds can affect ecosystems.

Zavarzin (1993) observed that endorhe-ic soda lakes have closed nutrient cycling, where the input of carbon and nitrogen into the ecosystem comes predominantly from CO2 and N2 fixation by photosynthetic cya-nobacteria. On other hand, the soda lakes are not entirely closed systems because of

sometimes vast populations of waterbirds (Brian et al. 1998; Sorokin et al. 2014).The soda pans of the Carpathian Basin are impor-tant breeding and stopover sites for African-Eurasian migratory waterbirds traversing the Black Sea and the Mediterranean flyway (Bo-ros et al. 2013). The aquatic communities of these habitats are strongly linked to water-birds in three ways: 1) soda pans with high zooplankton biomass attract planktivorous filter-feeding waterbirds (Boros et al. 2006a; Horváth et al. 2013); 2) wading shorebirds feed on less abundant nektonic and benthic invertebrates (Boros et al. 2006b, 2008b); and 3) large-bodied herbivorous bird spe-cies (e.g., geese) provide a significantly high external nutrient load of carbon (C), nitro-gen (N), and phosphorus (P), which causes hypertrophic conditions (Boros et al. 2008a) and may lead to net heterotrophy (Vörös et al. 2008).

The primary objective of this study was to identify the impact of waterbirds on the pro-cesses of production and decomposition in two characteristic Hungarian soda pans. Our

WaTErBird-mEdiaTEd ProduCTiViTy 389

hypothesis was that waterbirds have a sig-nificant impact on both planktonic produc-tion and respiration because of the nutrient loading associated with their presence. By selecting two different hypertrophic soda pans, one dominated by open water (80%) and one by macrophyte vegetation (90%), the spatial distribution of marshland macro-phytes was also taken into account as a back-ground environmental factor.

mEThods

Study Area

The pans were located in the Duna-Tisza Interflu-vial region of Hungary. The Zab-szék pan (46° 50' 5.09" N, 19° 10' 17.82" E) was selected as a prototypical turbid type (grayish-white water rich in suspended colloidal minerals), while the Sósér pan (46° 47' 18.62" N, 19° 8' 39.71" E) was selected to represent the colored type (brownish-black water rich in dissolved humic matters). The two pans are located 3 km apart. The pans are situ-ated in the groundwater discharge areas of a closed hy-drographic (endorheic) basin, in which groundwater inflow exceeds the surface-related watershed inflow and precipitation (Boros et al. 2013). We assume that nutri-ent input by precipitation, surface and groundwater in-flow were the same in both pans.

Submerged macrophytes were absent from the open waters of the pans, but both pans were surrounded by marshland vegetation dominated by saltmarsh bulrush (Bolboschoenus maritimus) and common reed (Phragmites australis) in varying spatial proportions. The extent of open water and marshland vegetation was estimated from field observations and remote-sensing databases extracted from Google Earth. The mapping procedure and spatial calculations were carried out using ArcMap (Environmental Systems Research Institute 2013). The selected pans differed in the spatial distribution of open water and macrophyte-covered areas: in 2014, the tur-bid Zab-szék pan (143 ha) consisted of 80% open water and 20% marshland vegetation, whereas the colored Sósér pan (51 ha) consisted of 90% marshland vegeta-tion and only 10% open water at average water-depth levels. As both pans are shallow with intermittent stand-ing water, they are incapable of supporting resident fish populations.

Field Measurements and Water Sampling

Water sampling was carried out monthly between January and December 2014 in the open waters of the pans. Water depth was measured at each sampling loca-tion, and underwater light conditions were measured with a Secchi disc (Secchi) except when the pan bottom was visible (Wetzel and Likens 1991). Water tempera-ture, pH and conductivity were measured using a WTW MultiLine P4 field instrument. The diel variation in wa-

ter temperature was determined using HOBO Pendant Temperature/Light Data Loggers (64K-UA-002-64). Sa-linity (Sal) was calculated from conductivity using the following formula: salinity (g/l) = 0.8 × conductivity (mS/cm) (Boros et al. 2014).

Laboratory Measurements

The concentration of colored dissolved organic matter (CDOM) was expressed as Pt (platina) units (mg Pt/l) using the method of Cuthbert and del Giorgio (1992). Acid-washed glassware was used for the analyses. Samples for total organic carbon (TOC) and dissolved organic carbon (DOC) analyses were acidified (to pH 2 with HCl) and bubbled to remove dissolved inorganic carbon. Next, the samples were filtered through a pre-combusted GF-5 glass fiber filter (pore size = 0.4 µm), and both TOC and DOC concentrations were measured using an Elementar High TOC (V.-Balogh et al. 2009, 2010). Particulate organic carbon (POC) was defined as the difference between TOC and DOC concentrations (POC = TOC – DOC).

For the determination of total nitrogen (TN), samples were digested by sodium hydroxide and potas-sium-persulfate in an autoclave at a temperature of 121 °C, and TN was measured with a Shimadzu UV-160A spectrophotometer (Eaton et al. 1995). The quantity of soluble reactive phosphorus (SRP) was determined from filtrates based on the method of Murphy and Riley (1962), while total phosphorus (TP) content was mea-sured spectrophotometrically after potassium-persul-fate digestion from unfiltered water samples (Menzel and Corwin 1965).

Total suspended solids (TSS) were determined grav-imetrically (Eaton et al. 1995) after filtration through cellulose acetate filters (pore size = 0.4 µm). For the chlorophyll a (Chl) measurement, 5-100 ml of each sample (depending on turbidity) was filtered through a GF-5 glass fiber filter, after which chlorophyll a concen-tration was determined spectrophotometrically with hot methanol extraction (Wetzel and Likens 1991). Algal-free suspended solids (TSS-Alg) concentration was cal-culated by assuming a 1:100 ratio between chlorophyll a concentration and the dry weight of phytoplankton (Reynolds 1984). Subtracting the latter from the total suspended solids (TSS) concentration yielded the algal-free suspended solids (TSS-Alg) concentration.

Planktonic Primary Production and Respiration

Planktonic primary production (Pro) and com-munity respiration (Res) were measured monthly using the ‘light-dark bottle’ method (Vollenweider 1969). The measurements began after ice melt from 1 February 2014 to 31 December 2014, except in July, when the pans were almost completely dry. Seasonal variation was monitored with two measurements per month during the vegetation period from March to October 2014, except in July and August, again be-cause of low water levels. To estimate production and respiration rates, samples were collected in 300 ml Karlsruhe bottles in the field. For each pan, three light bottles and three darkened bottles were placed

390 WaTErBirds

into the water along the shoreline. The bottles were fully submerged in an upside-down position, with the bottom of the bottles at the surface and the height of the Karlsruhe bottles extending beyond the euphotic depth of the ponds. Incubations lasted from 10 to 14 hr. Dissolved oxygen before and after the incubations was measured with a WTW Oxi 539 dissolved oxygen meter and a TriOxmatic 300 electrode. Daily produc-tion was calculated from the hourly values by assum-ing that Production(daily) = Production(hourly) × daylight hours. Daily respiration (for 24 hr) was cal-culated on the basis of the measured hourly respira-tion rates taking into account the diel variability of respiration as a consequence of temperature depen-dence (del Giorgio and le B. Williams 2005). Oxygen concentration data were transformed into carbon concentration (mg C/l) by applying a constant pa-rameter of 0.313.

Estimation of the Nutrient Loading of Waterbirds

The nutrient loading of birds was estimated by de-termining the abundance of waterbird populations and the nutrient content of their excrement. Waterbirds on the open water of the pans were counted with bin-oculars (8×42 and 10×42) and spotting scopes (30×75 and zoom 20-60×78) during daylight at 7-day intervals throughout 2014. Daily bird abundance (individu-als/ha) was calculated from the average of the weekly counts for each month. The contribution of waterbird populations to the daily nutrient load was estimated using daily net rates of C, N, and P excretion (Table 1). Daily carbon, nitrogen and phosphorus excretion data (g/day/individual) were modified by a species-residency time correction factor (RTF: residence time in hours on the pans during 24 hr) on the water surface based on observed diurnal and nocturnal activity. Daily net C, N, and P excrement data and waterbird species RTFs are listed in Table 2. The monthly total loading of waterbirds = Σ species (A × E × RTF × D), where A is the daily mean of abundance of waterbird species for each month, E is the daily net rate of C, N, and P in the excrement of each bird species, RTF is the daily residency time factor of each species on the open water and D is the number of days of each month. The annual summarized net C, N, and P loads were determined by summing the monthly average loads. Surface-related data were calculated as the sum of monthly loading volumes divided by the actual size of the open water of each month.

Statistical Analyses

We tested for relationships between waterbird-related nutrient loading and the investigated water variables with pairwise Spearman rank correlations. To reveal any possible relationship between mea-sured and estimated parameters, we also carried out a Principal Component Analysis (PCA). Basic statisti-cal analyses, graph preparation, PCA ordination with standardized scores and Spearman rank correlations were performed using Origin software (OriginLab 2015).

rEsulTs

Waterbird Population and Nutrient Loading

Both waterbird species richness and abundance were higher on the Zab-szék pan than on the Sósér pan, likely because of the difference in the size of the open water area. Altogether, 45 waterbird species were ob-served on the two pans, but only 14 of them were observed on the Sósér pan whereas only 44 of them were found on the Zab-szék pan (Table 2). Bird abundance and nutrient loading displayed notable seasonal variation, which was associated with peaks in the pas-sage of migratory birds. The abundance of waterbirds observed on the Zab-szék pan was highest in early spring (February-March), with a second peak (July) of post-breeding birds occurring prior to the drying out of the pans (Fig. 1A). On the Sósér pan, only moderate population peaks occurred in spring and summer (Fig. 1B). The high-est observed waterbird abundance was 171 individuals/ha on Zab-szék pan in Febru-ary, while there were only 26 individuals/ha on the Sósér pan at the beginning of July as the pan was drying out. The cumulative yearly sum of the monthly average number of waterbirds was two orders of magnitude higher and the yearly sum of the monthly average abundance was one order of magni-tude higher on the Zab-szék pan (55,453 in-dividuals; 385.10 individuals/ha) compared to Sósér pan (495 individuals; 41.24 indi-viduals/ha). The species composition of the waterbird community differed between the pans. Based on the number of individuals/ha, the primary species contributing to the allochthonous nutrient loads on the Zab-szék pan were the Greater White-fronted Goose, Greylag Goose, Eurasian Teal, Mal-lard, Eurasian Curlew, Caspian Gull, Yellow-legged Gull, and Black-headed Gull, while on the Sósér pan, the Northern Lapwing and the Black-headed Gull were the only abundant species (scientific names in Tables 1 and 2).

In 2014, the yearly total nutrient loading of waterbirds was estimated to be 7,582 mg C/m2/year, 1,221 mg N/m2/year and 197

WaTErBird-mEdiaTEd ProduCTiViTy 391

Tab

le 1

. Dai

ly n

et C

arbo

n (C

), N

itro

gen

(N),

and

Pho

spho

rus

(P)

excr

emen

t da

ta (

g/da

y/in

divi

dual

) an

d re

side

ncy

tim

e fa

ctor

s (R

TF:

res

iden

cy t

ime

hour

s on

the

pan

s du

ring

24

hou

rs)

of th

e ob

serv

ed w

ater

bird

spe

cies

. Int

erna

tion

al E

nglis

h an

d sc

ient

ific

nam

es a

re f

rom

the

Bri

tish

Orn

itho

logi

sts’

Uni

on (

2013

).

Wat

erbi

rd S

peci

esR

TF

C N

PSo

urce

Tun

dra/

Taig

a B

ean

Goo

seA

nser

faba

lis0.

69.

760.

490.

11O

láh

(20

03)

Gre

ylag

Goo

seA

nser

ans

er0.

69.

760.

490.

11O

láh

(20

03)

Gre

ater

Wh

ite-

fron

ted

Goo

seA

nser

alb

ifron

s0.

68.

600.

690.

08K

ear

(196

3)D

abbl

ing

duck

sA

nas

spp.

0.8

9.12

0.58

0.18

Man

ny

et a

l. (1

994)

; Olá

h (

2003

)D

ivin

g du

cks

Ayt

hya

and

Mer

gus

spp.

1.0

9.69

0.61

0.19

Man

ny

et a

l. (1

994)

; Olá

h (

2003

)G

reat

Egr

etA

rdea

alb

a1.

014

.50

1.38

3.78

Mar

ion

et a

l. (1

994)

Gre

y H

eron

Ard

ea c

iner

ea1.

014

.50

1.38

3.78

Mar

ion

et a

l. (1

994)

Gre

bes

Podi

ceps

spp

.1.

09.

690.

610.

19M

ann

y et

al.

(199

4); O

láh

(20

03)

Com

mon

Cra

ne

Gru

s gr

us0.

68.

403.

480.

58O

láh

(20

03)

Pied

Avo

cet

Rec

urvi

rost

ra a

vose

tta1.

05.

002.

160.

36O

láh

(20

03)

Bla

ck-w

inge

d St

iltH

iman

topu

s hi

man

topu

s1.

05.

002.

160.

36O

láh

(20

03)

Cur

lew

sN

umen

ius

spp.

0.6

3.00

1.30

0.22

Olá

h (

2003

)N

orth

ern

Lap

win

gVa

nellu

s va

nellu

s0.

62.

520.

650.

12O

láh

(20

03)

Plov

ers

Cha

radr

ius

spp.

1.0

3.00

0.93

0.11

Olá

h (

2003

)R

uff

Philo

mac

hus

pugn

ax0.

62.

520.

650.

12O

láh

(20

03)

San

dpip

ers

1C

alid

ris

spp.

1.0

3.00

0.93

0.11

Olá

h (

2003

)Sa

ndp

iper

s 2

Trin

ga s

pp.

1.0

4.20

1.08

0.20

Olá

h (

2003

)C

omm

on S

nip

eG

allin

ago

galli

nago

1.0

4.20

1.08

0.20

Olá

h (

2003

)B

lack

-hea

ded

Gul

lC

hroi

coce

phal

us r

idib

undu

s0.

63.

480.

360.

23G

ould

an

d Fl

etch

er (

1978

)M

edit

erra

nea

n G

ull

Lar

us m

elan

ocep

halu

s0.

63.

480.

360.

23B

oros

et a

l. (2

008a

)C

aspi

an G

ull

Lar

us c

achi

nnan

s0.

67.

680.

660.

62G

ould

an

d Fl

etch

er (

1978

)M

ew G

ull

Lar

us c

anus

0.6

4.32

0.48

0.30

Gou

ld a

nd

Flet

cher

(19

78)

Tern

sC

hlid

onia

s sp

p. a

nd S

tern

a sp

p.1.

04.

500.

600.

38B

oros

et a

l. (2

008a

)Ye

llow

-legg

ed G

ull

Lar

us m

icha

helli

s0.

67.

680.

660.

62G

ould

an

d Fl

etch

er (

1978

)

392 WaTErBirdsT

able

2. M

onth

ly a

vera

ge in

divi

dual

s an

d ab

unda

nce

of th

e ob

serv

ed b

ird

spec

ies

on th

e pa

ns in

201

4. I

nter

nati

onal

Eng

lish

and

scie

ntifi

c na

mes

are

fro

m th

e B

riti

sh O

rnit

holo

-gi

sts’

Uni

on (

2013

).

Wat

erbi

rd S

peci

esZ

ab-s

zék

Pan

(# in

divi

dual

s)Só

sér

Pan

(# in

divi

dual

s)Z

ab-s

zék

Pan

(# in

divi

dual

s/h

a)Só

sér

Pan

(# in

divi

dual

s/h

a)

Gre

ater

Wh

ite-

fron

ted

Goo

seA

nser

alb

ifron

s36

,930

256.

46G

reyl

ag G

oose

Ans

er a

nser

1,64

310

011

.41

8.33

Tun

dra/

Taig

a B

ean

Goo

seA

nser

faba

lis2

0.01

Red

-bre

aste

d G

oose

Bra

nta

rufic

ollis

50.

03N

orth

ern

Pin

tail

Ana

s ac

uta

910.

63N

orth

ern

Sh

ovel

erA

nas

clyp

eata

793

5.51

Eur

asia

n T

eal

Ana

s cr

ecca

5,81

840

.40

Eur

asia

n W

igeo

nA

nas

pene

lope

224

1.56

Mal

lard

Ana

s pl

atyr

hync

hos

2,72

318

.91

Gar

gan

eyA

nas

quer

qued

ula

430.

30G

adw

all

Ana

s st

repe

ra20

0.14

Com

mon

Poc

har

dA

ythy

a fe

rina

50.

03C

omm

on S

hel

duck

Tado

rna

tado

rna

151

1.05

Gre

at E

gret

Ard

ea a

lba

370.

26G

rey

Her

onA

rdea

cin

erea

10.

01C

omm

on C

ran

eG

rus

grus

600.

42C

omm

on S

andp

iper

Act

itis

hypo

leuc

os2

0.01

Dun

linC

alid

ris

alpi

na62

91

4.37

0.08

Cur

lew

San

dpip

erC

alid

ris

ferr

ugin

ea4

0.03

Lit

tle

Stin

tC

alid

ris

min

uta

151

0.10

0.08

Tem

min

ck’s

Sti

nt

Cal

idri

s te

mm

inck

ii3

30.

020.

25K

enti

sh P

love

rC

hara

driu

s al

exan

drin

us8

0.06

Lit

tle

Rin

ged

Plov

erC

hara

driu

s du

bius

200.

14C

omm

on R

inge

d Pl

over

Cha

radr

ius

hiat

icul

a8

240.

062.

00C

omm

on S

nip

eG

allin

ago

galli

nago

30.

000.

25B

lack

-win

ged

Stilt

Him

anto

pus

him

anto

pus

2115

0.15

1.25

Bro

ad-b

illed

San

dpip

erL

imic

ola

falc

inel

lus

10.

01E

uras

ian

Cur

lew

Num

eniu

s ar

quat

a16

31.

13W

him

brel

Num

eniu

s ph

aeop

us1

0.01

Ruf

fPh

ilom

achu

s pu

gnax

2,47

512

617

.19

10.5

0Pi

ed A

voce

tR

ecur

viro

stra

avo

setta

709

34.

920.

25C

omm

on T

ern

Ster

na h

irun

do1

0.01

Spot

ted

Red

shan

kTr

inga

ery

thro

pus

334

312.

322.

58W

ood

San

dpip

erTr

inga

gla

reol

a75

580.

524.

83C

omm

on G

reen

shan

kTr

inga

neb

ular

ia11

0.08

Mar

sh S

andp

iper

Trin

ga s

tagn

atili

s5

30.

030.

25

WaTErBird-mEdiaTEd ProduCTiViTy 393

mg P/m2/year for the Zab-szék pan, and 1,376 mg C/m2/year, 226 mg N/m2/year

and 54 mg P/m2/year for the Sósér pan. The yearly total nutrient loads and the sea-sonal dynamics of waterbird abundance show that the monthly mean carbon, nitro-gen and phosphorus loading by waterbirds was six, five, and four times higher in the Zab-szék pan, respectively, than in the Só-sér pan (Table 3; Fig. 1B, 1C).

Physical and Chemical Conditions of the Pans

For both pans, the salinity was within the hyposaline range (5.6-5.7 g/l) and the pH was alkaline (> 9). Both pans were very shallow with water depths ranging from 2 to 50 cm, and a monthly period of drying out from the middle of July through Au-gust. The transparency, as measured with Secchi, was very low in the turbid Zab-szék pan; the mean transparency for the Zab-szék pan was 1.9 cm, with a minimum of 0.3 cm, due to the high concentration of algal-free suspended solids. However, transpar-ency was also low in the colored Sósér pan, with a much lower concentration of algal-free suspended solids and a mean CDOM concentration an order of magnitude high-er than in the Zab-szék pan (Table 3).

There was a strong positive correlation between CDOM and DOC (r = 0.862), and, on the basis of the DOC concentra-tions, both pans (Sósér in particular) were polyhumic (> 16 mg C/l). The proportion of POC was 32% for the Zab-szék pan and only 9% for the Sósér pan. The mean total nitrogen concentration was higher (2.6x) in the Sósér pan compared to the Zab-szék pan. Conversely, both mean total and solu-ble reactive phosphorus concentration was higher (3.6x and 4.3x, respectively) in the Zab-szék pan compared to the Sósér pan (Table 3).

Planktonic Primary Production and Respi-ration

Mean chlorophyll a concentration was two times higher in the Zab-szék pan than T

able

2. (

cont

inue

d) M

onth

ly a

vera

ge in

divi

dual

s an

d ab

unda

nce

of th

e ob

serv

ed b

ird

spec

ies

on th

e pa

ns in

201

4. I

nter

nati

onal

Eng

lish

and

scie

ntifi

c na

mes

are

fro

m th

e B

riti

sh

Orn

itho

logi

sts’

Uni

on (

2013

).

Wat

erbi

rd S

peci

esZ

ab-s

zék

Pan

(# in

divi

dual

s)Só

sér

Pan

(# in

divi

dual

s)Z

ab-s

zék

Pan

(# in

divi

dual

s/h

a)Só

sér

Pan

(# in

divi

dual

s/h

a)

Com

mon

Red

shan

kTr

inga

tota

nus

770.

53N

orth

ern

Lap

win

gVa

nellu

s va

nellu

s1,

355

209.

411.

67W

his

kere

d Te

rnC

hlid

onia

s hy

brid

a3

0.02

Bla

ck T

ern

Chl

idon

ias

nige

r5

0.03

Bla

ck-h

eade

d G

ull

Chr

oico

ceph

alus

rid

ibun

dus

749

107

5.20

8.92

Mew

Gul

lL

arus

can

us4

0.03

Med

iter

ran

ean

Gul

lL

arus

mel

anoc

epha

lus

30.

02Ye

llow

-legg

ed G

ull &

Cas

pian

Gul

lL

arus

mic

hahe

llis

& L

arus

cac

chin

nans

226

1.57

TO

TA

L55

,453

495

385.

1041

.24

394 WaTErBirds

in the Sósér pan. The seasonal dynamics of chlorophyll a concentration was also differ-ent. In the Zab-szék pan, the annual peak occurred in spring, while in the Sósér pan it was observed in summer (Fig. 1D).

There was a significant correlation (r = 0.596) between the monthly Pro and Res of the pans, and the rate of respiration was high-er than the rate of production for both pans. The yearly Pro/Res ratio was 0.85 in the Zab-

szék pan but only 0.29 in the Sósér pan. The estimated mean monthly production rates (Pro: mg C/m2/month) were twice as high in the Zab-szék pan (Fig. 1E)—where the nutri-ent loading of waterbirds was also higher (C: 6x; N: 5x; P: 4x)—compared to the Sósér pan, while the mean monthly respiration rates (Res: mg C/m2/month) were similar (Table 3).

The seasonal dynamics of production and respiration also differed between the

Figure 1. Seasonal dynamics of waterbird abundance (A), nutrient loading of waterbird excrement (B and C), chlo-rophyll a concentration (D), planktonic production (E), and respiration (F) on the investigated pans (Zab-szék and Sósér pans) in 2014.

WaTErBird-mEdiaTEd ProduCTiViTy 395

Tab

le 3

. Bas

ic s

tati

stic

al s

umm

ary

of th

e so

da p

ans

in 2

014.

Para

met

erA

bbre

viat

ion

Num

ber

Mea

nM

inim

umM

axim

um

Zab-

szék

Pan

Wat

er d

epth

(cm

)Z

1813

.00

2.00

40.0

0Se

cch

i dis

c tr

ansp

aren

cy (

cm)

Secc

hi

181.

900.

304.

00A

lgal

-free

sus

pen

ded

solid

s (m

g/l)

TSS

-Alg

182,

182.

8011

1.20

7,63

7.90

Salin

ity

(g/l

)Sa

l18

5.60

1.70

17.0

0pH

pH18

9.60

8.90

10.2

0C

olor

ed d

isso

lved

org

anic

mat

ter

(mg

Pt/l

)C

DO

M17

470.

4019

9.30

926.

10D

isso

lved

org

anic

car

bon

(m

g/l)

DO

C18

69.1

229

.43

217.

19To

tal o

rgan

ic c

arbo

n (

mg/

l)T

OC

1814

0.67

34.1

540

7.29

Tota

l nit

roge

n (

µg/l

)T

N14

8,88

5.12

2,02

6.39

27,4

41.1

6So

lubl

e re

acti

ve p

hos

phor

us (

µg/l

)SR

P14

3,04

4.00

207.

3015

,239

.07

Tota

l ph

osph

orus

(µg

/l)

TP

137,

170.

082,

012.

4717

,703

.67

Ch

loro

phyl

l a (

µg/l

)C

hl

1875

1.95

6.06

2,75

0.84

Plan

kton

ic g

ross

pri

mar

y pr

oduc

tion

(m

g C

/m2 /

mon

th)

Pro

143,

941.

310.

0016

,092

.38

Plan

kton

ic r

espi

rati

on (

mg

C/m

2 /m

onth

)R

es14

7,23

6.42

920.

9323

,356

.43

Car

bon

load

ing

of w

ater

bird

exc

rem

ent (

mg

C/m

2 /m

onth

)C

1263

1.80

66.3

62,

679.

31N

itro

gen

load

ing

of w

ater

bird

exc

rem

ent (

mg

N/m

2 /m

onth

)N

1210

1.75

17.9

144

5.18

Phos

phor

us lo

adin

g of

wat

erbi

rd e

xcre

men

t (m

g P/

m2 /

mon

th)

P12

16.4

53.

9456

.30

396 WaTErBirds

Tab

le 3

. (co

ntin

ued)

Bas

ic s

tati

stic

al s

umm

ary

of th

e so

da p

ans

in 2

014.

Para

met

erA

bbre

viat

ion

Num

ber

Mea

nM

inim

umM

axim

um

Sósé

r P

anW

ater

dep

th (

cm)

Z18

18.0

55.

5050

.00

Secc

hi d

isc

tran

spar

ency

(cm

)Se

cch

i18

7.20

1.50

20.0

0A

lgal

-free

sus

pen

ded

solid

s (m

g/l)

TSS

-Alg

1897

.40

0.00

470.

70Sa

linit

y (g

/l)

Sal

185.

701.

3012

.90

pHpH

189.

308.

4010

.10

Col

ored

dis

solv

ed o

rgan

ic m

atte

r (m

g Pt

/l)

CD

OM

175,

388.

701,

148.

409,

644.

70D

isso

lved

org

anic

car

bon

(m

g/l)

DO

C18

437.

2595

.38

988.

49To

tal o

rgan

ic c

arbo

n (

mg/

l)T

OC

1847

9.20

98.6

11,

138.

15To

tal n

itro

gen

(µg

/l)

TN

1422

,725

.31

3,52

0.12

62,6

61.4

6So

lubl

e re

acti

ve p

hos

phor

us (

µg/l

)SR

P14

710.

8933

.89

2,52

1.83

Tota

l ph

osph

orus

(µg

/l)

TP

142,

015.

7829

2.29

3,97

9.60

Ch

loro

phyl

l a (

µg/l

)C

hl

1836

9.12

1.16

3,14

4.48

Plan

kton

ic g

ross

pri

mar

y pr

oduc

tion

(m

g C

/m2 /

mon

th)

Pro

141,

872.

050.

0014

,645

.27

Plan

kton

ic r

espi

rati

on (

mg

C/m

2 /m

onth

)R

es14

8,03

9.55

207.

6545

,162

.81

Car

bon

load

ing

of w

ater

bird

exc

rem

ent (

mg

C/m

2 /m

onth

)C

1211

4.65

0.00

349.

32N

itro

gen

load

ing

of w

ater

bird

exc

rem

ent (

mg

N/m

2 /m

onth

)N

1218

.80

0.00

68.6

1Ph

osph

orus

load

ing

of w

ater

bird

exc

rem

ent (

mg

P/m

2 /m

onth

)P

124.

510.

0013

.46

WaTErBird-mEdiaTEd ProduCTiViTy 397

pans. In the Zab-szék pan, maximum pro-duction and respiration both occurred in the spring, during February-April (Fig. 1E), when the nutrient loading of waterbirds was also at its peak (Fig. 1B). In contrast, in the Sósér pan, both production and respiration peaked in August (Fig. 1F), due to the pres-ence of a dense population of blue-green algae, with an extremely high maximum chlorophyll a concentration in the shallow water occurring after a month-long dry pe-riod. This peak was of a brief duration, and production was almost zero from the middle of September to the end of the year.

The estimated annual planktonic pro-duction was 2.8 times higher in the Zab-szék pan (63,468 mg C/m2/year) than in the Só-sér pan (22,796 mg C/m2/year). In contrast to production, annual planktonic respira-tion was more or less the same in the two pans (Zab-szék pan: 74,748 mg C/m2/year; Sósér pan: 77,460 mg C/m2/year).

Principal Component Analysis of the Vari-ables

The first four principal components ex-plain 85.67% of the variance, the remaining components each contribute 4% or less, and several variables are highly correlated. Ac-cording to the PCA ordination and pairwise correlations, most of the physical and chemi-cal variables related to the first, second or third component, and they varied together in three groups. The first group was com-posed of T, Sal, pH, SRP and TP, the second comprised Z, Secchi and TN, and the third included Pt, TOC and DOC. TSS-Alg varied separately along the second component. The C, N, and P loading of waterbirds varied closely together with Chl and Pro along the second component, while only Res varied mostly with the fourth component, and it had no significant correlation with any mea-sured variables (Fig. 2).

The C loading of waterbirds was signifi-cantly and positively correlated with the TSS-Alg, TP, SRP and Pro variables, and nega-tively correlated with the Secchi, CDOM and DOC variables. The N loading of waterbirds significantly and positively correlated with

the TSS-Alg, pH, CDOM, TP, SRP, Chl and Pro variables, and negatively correlated with the Z, Secchi and DOC variables. The P load-ing of waterbirds significantly and positively correlated with the TSS-Alg, pH, TP, SRP, Chl and Pro variables, and negatively corre-lated with the Secchi, CDOM and DOC vari-ables (Table 4).

disCussion

Both the turbid and colored pans were considered to be hypertrophic based on the concentration of inorganic plant nutri-ents and chlorophyll a (Organization for Economic Cooperation and Development 1982), but the depth-integrated production level was very low compared to the chloro-phyll a concentration, and the gross primary production and respiration ratio was smaller than 1 for both pans. Although some studies have found that a Pro/Res ratio below 1 is common at different trophic states in tem-perate zone lakes (del Giorgio and Peters 1994; Hanson et al. 2003; Urabe et al. 2005; Cornell and Klarer 2008), we found that a Pro/Res < 0.5 may be common in soda pans. Del Giorgio and Peters (1993) demonstrat-ed that the Pro/Res ratio varies with trophic status. Oligotrophic lakes are considered to be net heterotrophic, with chlorophyll a concentrations of 17 µg/l set as the divid-ing point. Eutrophic and hypertrophic lakes are considered to be net autotrophic due to the higher levels of primary productivity and lower allochthonous organic matter. Our results are inconsistent with these findings despite the fact that the average chlorophyll a concentration of the pans was an order of magnitude higher (means of turbid: 752 µg/l and colored: 369 µg/l) than the sug-gested limit (17 µg/l) of heterotrophy; there was no question that the pans were net het-erotrophic.

Waterbird-related nutrient loading in inland waters can vary widely (Manny et al. 1994; Post et al. 1998; Hahn et al. 2008). How-ever, in the current study phosphorus and nitrogen loads were found to be three and six times higher, respectively, compared to a

398 WaTErBirds

prototypical example of guanotrophication (Manny et al. 1994) on Wintergreen Lake, Michigan, USA, where the estimated load-ing of waterbirds was 297 kg/ha/year for C, 19 kg/ha/year for N and 6 kg/ha/year for P. Thus, our study represents an example of extreme guanotrophication in a continental environment; guanotrophication may be a consequence of the relatively small size of the pans and the high abundance and diver-sity of waterbirds present.

Some previous studies (Unckless and Ma-karewicz 2007; Özbay 2015) did not find any significant differences in the plant nutrient pool and concentration of chlorophyll a at a 1.5- to 2-m water depth when treated with bird excrement, indicating that feces and

associated nutrients settled quickly to the bottom sediments. By contrast, the two soda pans of the current study were very shallow, with a mean depth of 0.13 m and 0.19 m, respectively, which allowed for the continu-ous mixing of the water column. Therefore, feces and associated nutrients did not settle to the bottom sediments, which explains the significant positive correlation among the nutrient loading of waterbirds, TP and SRP.

The current study also demonstrated that the nutrient loading of waterbirds has an im-portant role in soda pan phosphorus cycling, which was already suggested in a bird-free control investigation (Boros et al. 2008a). Our hypothesis was supported, as waterbirds significantly affected planktonic gross pri-

Figure 2. Presentation of the principal component analysis (PCA) of Carbon (C), Nitrogen (N), and Phosphorus (P) loading of waterbird excrement and the environmental parameters. Dashed lined ellipse symbolizes that nutri-ent (C, N, P) loading of waterbirds closely varied together with Chl and Pro along the second component (see in Table 3 for abbreviations and units of variables).

WaTErBird-mEdiaTEd ProduCTiViTy 399

mary production in the studied soda pans, presumably as a consequence of the biologi-cal availability of their excrement. Despite the high nutrient (N, P) concentrations and high phytoplankton biomass, depth-inte-grated primary production was low due to limited light availability caused by extremely high turbidity and high CDOM concentra-tion. In contrast to production, there was no significant correlation between the nu-trient loading of waterbirds and planktonic respiration. Thus, this aspect of the hypoth-esis was not supported. The low Pro/Res in the colored pan was presumably caused by a high dissolved organic carbon concentra-tion (polyhumic). A possible explanation for the difference of Pro/Res between the turbid and colored pans is variation in the decomposition of the bird excrement and surrounding macrophytes.

aCknoWlEdgmEnTs

Data collection was supported by the Kiskunság National Park Directorate with a legal permission for conducting our research. We are grateful to the staff of the national park, who helped in our fieldwork, and to Naturglobe Ltd. for professional, technical and man-agement support. Boglárka Somogyi was sponsored by the Bolyai János research fellow of the Hungarian Acad-emy of Sciences. We thank Anett Rozália Kelemen and Balázs Németh for the laboratory measurements, and Verner Peter Bingman and Károly Pálffy for the English editing. Our methods meet all ethical guidelines for the use of wild birds in research, as stipulated by the stan-dards and policies of Hungary.

liTEraTurE CiTEd

Boros, E., Sz. Bánfi and L. Forró. 2006a. Anostracans and microcrustaceans as potential food sources of waterbirds on sodic pans of the Hungarian plain. Hydrobiologia 567: 341-349.

Boros, E., S. Andrikovics, B. Kiss and L. Forró. 2006b. Feeding ecology of migrating waders (Charadrii) at sodic-alkaline pans, in the Carpathian Basin. Bird Study 53: 86-91.

Boros, E., T. Nagy, Cs. Pigniczki, L. Kotymán, K. V.-Balogh and L. Vörös. 2008a. The effect of aquatic birds on the nutrient load and water quality of soda pans in Hungary. Acta Zoologica Academiae Scientiarum Hungaricae 54 (Supplement 1): 207-224.

Boros, E., L. Forró, G. Gere, O. Kiss, L. Vörös and S. Andrikovics. 2008b. The role of aquatic birds in the regulation of trophic relationships of conti-

Tab

le 4

. Spe

arm

an’s

ran

k co

rrel

atio

n co

effi

cien

ts b

etw

een

Car

bon

(C),

Nit

roge

n (N

), a

nd P

hosp

horu

s (P

) lo

adin

g of

wat

erbi

rd e

xcre

men

t and

phy

sica

l, ch

emic

al, a

nd b

iolo

gica

l va

riab

les

(***

P <

0.0

01, *

* P

< 0.

01, *

P <

0.0

5).

Car

bon

Loa

din

g of

Wat

erbi

rds

Nit

roge

n L

oadi

ng

of W

ater

bird

sPh

osph

orus

Loa

din

g of

Wat

erbi

rds

Vari

able

Cor

rela

tion

co

effi

cien

tVa

riab

leC

orre

lati

on

coef

fici

ent

Vari

able

Cor

rela

tion

co

effi

cien

t

Secc

hi d

isc

tran

spar

ency

−0.6

53**

*W

ater

dep

th−0

.485

*Se

cch

i dis

c tr

ansp

aren

cy−0

.748

***

Alg

al-fr

ee s

uspe

nde

d so

lids

+0.8

13**

*Se

cch

i dis

c tr

ansp

aren

cy−0

.705

***

Alg

al-fr

ee s

uspe

nde

d so

lids

+0.8

57**

*C

olor

ed d

isso

lved

org

anic

mat

ter

−0.4

91*

Alg

al-fr

ee s

uspe

nde

d so

lids

+0.8

23**

*pH

+0.4

95*

Dis

solv

ed o

rgan

ic c

arbo

n−0

.553

**pH

+0.6

08**

Col

ored

dis

solv

ed o

rgan

ic m

atte

r−0

.549

**To

tal p

hos

phor

us+0

.556

*C

olor

ed d

isso

lved

org

anic

mat

ter

+0.6

06**

Dis

solv

ed o

rgan

ic c

arbo

n−0

.416

*So

lubl

e re

acti

ve p

hos

phor

us+0

.535

*D

isso

lved

org

anic

car

bon

−0.4

20*

Tota

l ph

osph

orus

+0.6

92**

*Pl

ankt

onic

gro

ss p

rim

ary

prod

ucti

on+0

.468

*To

tal p

hos

phor

us+0

.654

**So

lubl

e re

acti

ve p

hos

phor

us+0

.642

**So

lubl

e re

acti

ve p

hos

phor

us+0

.590

**C

hlo

roph

yll a

+0.4

35*

Ch

loro

phyl

l a+0

.427

*Pl

ankt

onic

gro

ss p

rim

ary

prod

ucti

on+0

.635

**Pl

ankt

onic

gro

ss p

rim

ary

prod

ucti

on+0

.590

**

400 WaTErBirds

nental soda pans in Hungary. Acta Zoologica Aca-demiae Scientiarum Hungaricae 54 (Supplement 1): 189-206.

Boros, E., Z. Ecsedi and J. Oláh (Eds.). 2013. Ecol-ogy and management of soda pans in the Carpath-ian Basin. Hortobágy Environmental Association, Balmazújváros, Hungary.

Boros, E., Zs. Horváth, G. Wolfram and L. Vörös. 2014. Salinity and ionic composition of the shal-low astatic soda pans in the Carpathian Basin. Annales de Limnologie - International Journal of Limnology 50: 59-69.

Brian, E. J., W. D. Grant, A. W. Duckworth and G. G. Owenson. 1998. Microbial diversity of soda lakes. Extremophiles 2: 191-200.

British Ornithologists’ Union. 2013. The British list: a checklist of birds of Britain, 8th ed. Ibis 155: 635-676.

Cornell, L. P. and D. M. Klarer. 2008. Patterns of dis-solved oxygen, productivity and respiration in Old Woman Creek Estuary, Erie County, Ohio during low and high water conditions. Ohio Journal of Science 108: 31-43.

Cuthbert, I. D. and P. del Giorgio. 1992. Toward a standard method of measuring colour in freshwa-ter. Limnology and Oceanography 37: 1319-1326.

del Giorgio, P. A. and R. H. Peters. 1993. Balance be-tween phytoplankton production and plankton respiration in lakes. Canadian Journal of Fisheries and Aquatic Sciences 50: 282-289.

del Giorgio, P. A. and R. H. Peters. 1994. Patterns in planktonic P:R ratios in lakes: influence of lake trophy and dissolved organic C. Limnology and Oceanography 39: 772-787.

del Giorgio, P. A. and P. J. le B. Williams. 2005. Res-piration in aquatic ecosystems. Oxford University Press, Oxford, U.K.

Eaton, A. D., L. S. Clesceri and A. E. Greenberg. 1995. Standard methods for the examination of water and wastewater, 19th ed. American Public Health Association, Washington, D.C.

Environmental Systems Research Institute (ESRI). 2013. ESRI ArcMap GIS software v. 10.2 for Desk-top, ESRI, Redlands, California. http://www.esri.com, accessed 30 July 2013.

Gould, D. J. and M. R. Fletcher. 1978. Gull droppings and their effects on water quality. Water Research 12: 665-762.

Green, A. J. and J. Elmberg. 2013. Ecosystem services provided by waterbirds. Biological Reviews 89: 105-122.

Hahn, S., S. Bauer and M. Klaassen. 2008. Quantifica-tion of allochthonous nutrient input into freshwa-ter bodies by herbivorous waterbirds. Freshwater Biology 53: 181-193.

Hanson, P. C., D. L. Bade, S. R. Carpenter and T. K. Kratz. 2003. Lake metabolism: relationships with dissolved organic carbon and phosphorus. Lim-nology and Oceanography 48: 1112-1119.

Horváth, Zs., Cs. F. Vad, L. Vörös and E. Boros. 2013. The keystone role of anostracans and copepods in

European soda pans during the spring migration of waterbirds. Freshwater Biology 58: 430-440.

Kear, J. 1963. The agricultural importance of wild goose droppings. Pages 72-77 in The Wildfowl Trust Fourteenth Annual Report, Slimbridge, U.K.

Manny, B. A., W. C. Johnson and R. G. Wetzel. 1994. Nutrient additions by waterfowl to lakes and reser-voirs: predicting their effects on productivity and water quality. Hydrobiologia 279-280: 121-132.

Marion, L., P. Clergeau, L. Brient and G. Bertru. 1994. The importance of avian-contributed nitro-gen (N) and phosphorus (P) to Lake Grand-Lieu, France. Hydrobiologia 279-280: 133-147.

Menzel, D. H. and N. Corwin. 1965. The measure-ment of total phosphorus in seawater based on the liberation of organically bound fractions by persulfate oxidation. Limnology and Oceanogra-phy 10: 280-282.

Murphy, J. and J. P. Riley. 1962. A modified single solution method for the determination of phos-phate in natural waters. Analytica Chimica Acta 27: 31-36.

Oláh, J. 2003. Waterbird trophic guilds. Hungarian Waterfowl Publication 10: 381-423. (In Hungarian with English summary).

Organization for Economic Cooperation and De-velopment (OECD). 1982. Eutrophication of waters. Monitoring, assessment and control. Un-published report, OECD Cooperative Programme on Monitoring of Inland Waters (Eutrophication Control), Environment Directorate, OECD, Paris, France.

OriginLab. 2015. Origin: data analysis and graphing software v. 9.2. OriginLab, Northampton, Massa-chusetts. http://www.originlab.com, accessed 24 October 2014.

Özbay, H. 2015. The effects of bird exclusion on the chemical and biological characteristics of a soda lake. International Journal of Aquatic Science 6: 3-10.

Post, D. M., J. P. Taylor, J. F. Kitchell, M. H. Olson, D. E. Schindler and B. R. Herwig. 1998. The role of migratory waterfowl as nutrient vectors in a man-aged wetland. Conservation Biology 12: 910-920.

Reynolds, C. S. 1984. The ecology of freshwater phy-toplankton. Cambridge University Press, Cam-bridge, U.K.

Sorokin, D. Y., T. Berben, E. Denise, M. L. Overmars, C. D. Vavourakis and G. Muyzer. 2014. Microbi-al diversity and biogeochemical cycling in soda lakes. Extremophiles 18: 791-809.

Unckless, R. L. and J. C. Makarewicz. 2007. The impact of nutrient loading from Canada Geese (Branta canadensis) on water quality, a mesocosm approach. Hydrobiologia 586: 393-401.

Urabe, J., T. Yoshida, T. B. Gurung, T. Sekino, N. Tsugeki, K. Nozaki, M. Maruo, E. Nakayama and M. Nakanishi. 2005. The production-to-respira-tion ratio and its implication in Lake Biwa, Japan. Ecological Research 20: 367-375.

WaTErBird-mEdiaTEd ProduCTiViTy 401

V.-Balogh, K., B. Németh and L. Vörös. 2009. Specific attenuation coefficients of optically active sub-stances and their contribution to the underwa-ter ultraviolet and visible light climate in shallow lakes and ponds. Hydrobiologia 632: 91-105.

V.-Balogh, K., B. Németh and L. Vörös. 2010. Szerves-anyagok magyarországi fehér vizű szikes tavakban. Acta Biologica Debrecina, Supplementum Oeco-logica Hungarica 22: 75-86. (In Hungarian).

Vollenweider, R. A. 1969. A manual on methods for measuring primary production in aquatic envi-

ronments. IBP Handbook No. 12, Blackwell Sci-entific Publications, Oxford, U.K.

Vörös, L., B. Somogyi and E. Boros. 2008. Birds cause net heterotrophy in shallow lakes. Acta Zoologica Academiae Scientiarum Hungaricae 54 (Supple-ment 1): 23-34.

Wetzel, R. G. and G. E. Likens. 1991. Limnological anal-ysis, 2nd ed. Springer-Verlag, New York, New York.

Zavarzin, G. A. 1993. Epicontinental soda lakes as probable relict biotopes of terrestrial biota forma-tion. Microbiology 62: 473-479.